@inproceedings{BilalKancharanaMayeretal.2022, author = {Bilal, M{\"u}henad and Kancharana, Sunil and Mayer, Christian and Pfaller, Daniel and Koval, Leonid and Bregulla, Markus and Cupek, Rafal and Ziębiński, Adam}, title = {High Resolution Mask R-CNN-based Damage Detection on Titanium Nitride Coated Milling Tools for Condition Monitoring by using a New Illumination Technique}, volume = {vol. 5: VISAPP}, booktitle = {Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications}, editor = {Farinella, Giovanni Maria and Radeva, Petia and Bouatouch, Kadi}, publisher = {SciTePress}, address = {Set{\´u}bal}, isbn = {978-989-758-555-5}, issn = {2184-4321}, doi = {https://doi.org/10.5220/0010781800003124}, pages = {305 -- 314}, year = {2022}, abstract = {The implementation of intelligent software in the manufacturing industry is a technology of growing importance and has highlighted the need for improvement in automatization, production, inspection, and quality assurance. An automated inspection system based on deep learning methods can help to enhance inspection and provide a consistent overview of the production line. Camera-based imaging systems are among the most widely used tools, replacing manual industrial quality control tasks. Moreover, an automatized damage detection system on milling tools can be employed in quality control during the coating process and to simplify measuring tool life. Deep Convolutional Neural Networks (DCNNs) are state-of-the-art methods used to extract visual features and classify objects. Hence, there is great interest in applying DCNN in damage detection and classification. However, training a DCNN model on Titanium-Nitride coated (TiN) milling tools is extremely challenging. Due to the coating, the optical properties such as reflection and light scattering on the milling tool surface make image capturing for computer vision tasks quite challenging. In addition to the reflection and scattering, the helical-shaped surface of the cutting tools creates shadows, preventing the neural network from efficient training and damage detection. Here, in the context of applying an automatized deep learning-based method to detect damages on coated milling tools for quality control, the light has been shed on a novel illumination technique that allows capturing high-quality images which makes efficient damage detection for condition monitoring and quality control reliable. The method is outlined along with results obtained in training a ResNet 50 and ResNet 101 model reaching an overall accuracy of 83\% from a dataset containing bounding box annotated damages. For instance and semantic segmentation, the state-of-the-art framework Mask R-CNN is employed.}, language = {en} } @thesis{Kancharana2022, author = {Kancharana, Sunil}, title = {Image based damage detection of milling tools}, publisher = {Technische Hochschule Ingolstadt}, address = {Ingolstadt}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:573-33653}, pages = {ii, 72}, year = {2022}, abstract = {Tool sharpness is critical in the production process because the tool's health determines the quality and precision of the workpiece. If the tool is damaged but the damage magnitude is minor, it is preferable to regrind the tool rather than purchasing a new one. This step aids in lowering the company's overall tool accounting costs. However, the complex helical shaped geometry of the tool along with the highly reflective optical properties on the surface makes the tool regrinding a challenging task. The existing state-of-the-art machines that perform regrinding for the optical critical milling tools have a number of drawbacks. First and foremost would be the time. Even with the latest technology, it takes around 15-20 minutes to process the entire tool for damage detection and the other major setback is the cost of the equipment. The current thesis provides a novel approach to detect the damages of the tools using tool images that are taken under a controlled experimental setup which is relatively cheaper and faster than the current state of art machines. A homogeneous illumination has been provided on the optical critical milling tools to obtain high-quality images. These images are later processed through classical image processing algorithms to determine the damaged regions of the tool. Different methods such as image segmentation method where masking is performed on the tool image using an automatic thresholding parameter. Due to innovative light source simplified methods such as mean subtraction where the image mean is subtracted from the original image to obtain the anomalies, contour-box scan method in which the tool edge contour is obtained and then the regions beside the tool edge are examined, the slope method where the tool edge slope is studied to determine the broken edges are implemented for damage detection and the challenges encountered by each method due the optical critical properties of the tool and the drawbacks for each method are outlined.}, language = {en} } @inproceedings{BilalMayerKancharanaetal.2022, author = {Bilal, M{\"u}henad and Mayer, Christian and Kancharana, Sunil and Bregulla, Markus and Cupek, Rafal and Ziębiński, Adam}, title = {Damage Detection of Coated Milling Tools Using Images Captured by Cylindrical Shaped Enclosure Measurement Setup}, booktitle = {Advances in Computational Collective Intelligence. 14th International Conference, ICCCI 2022, Hammamet, Tunisia, September 28-30, 2022, Proceedings}, editor = {Bădică, Costin and Treur, Jan and Benslimane, Djamal and Hnatkowska, Bogumiła and Kr{\´o}tkiewicz, Marek}, publisher = {Springer}, address = {Cham}, isbn = {978-3-031-16209-1}, doi = {https://doi.org/10.1007/978-3-031-16210-7_21}, pages = {264 -- 272}, year = {2022}, language = {en} } @inproceedings{BilalKancharanaMayeretal.2022, author = {Bilal, M{\"u}henad and Kancharana, Sunil and Mayer, Christian and Bregulla, Markus and Ziębiński, Adam and Cupek, Rafal}, title = {Image-based damage detection on TiN-coated milling tools by using a multi-light scattering illumination technique}, booktitle = {Fourteenth International Conference on Machine Vision (ICMV 2021)}, editor = {Osten, Wolfgang and Nikolaev, Dmitry and Zhou, Jianhong}, publisher = {SPIE}, address = {Bellingham}, isbn = {978-1-5106-5045-9}, doi = {https://doi.org/10.1117/12.2623140}, year = {2022}, language = {en} }